本文整理汇总了C++中STM2函数的典型用法代码示例。如果您正苦于以下问题:C++ STM2函数的具体用法?C++ STM2怎么用?C++ STM2使用的例子?那么, 这里精选的函数代码示例或许可以为您提供帮助。
在下文中一共展示了STM2函数的12个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的C++代码示例。
示例1: n2fv_4
static void n2fv_4(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
INT i;
const R *xi;
R *xo;
xi = ri;
xo = ro;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T1, T2, T4, T5;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
T4 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
T5 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
{
V T3, T7, T6, T8;
T3 = VSUB(T1, T2);
T7 = VADD(T1, T2);
T6 = VSUB(T4, T5);
T8 = VADD(T4, T5);
{
V T9, Ta, Tb, Tc;
T9 = VSUB(T7, T8);
STM2(&(xo[4]), T9, ovs, &(xo[0]));
Ta = VADD(T7, T8);
STM2(&(xo[0]), Ta, ovs, &(xo[0]));
Tb = VFMAI(T6, T3);
STM2(&(xo[6]), Tb, ovs, &(xo[2]));
STN2(&(xo[4]), T9, Tb, ovs);
Tc = VFNMSI(T6, T3);
STM2(&(xo[2]), Tc, ovs, &(xo[2]));
STN2(&(xo[0]), Ta, Tc, ovs);
}
}
}
}
示例2: n2bv_2
static void n2bv_2(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T1, T2, T3, T4;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
T3 = VSUB(T1, T2);
STM2(&(xo[2]), T3, ovs, &(xo[2]));
T4 = VADD(T1, T2);
STM2(&(xo[0]), T4, ovs, &(xo[0]));
STN2(&(xo[0]), T4, T3, ovs);
}
}
示例3: n2bv_6
static void n2bv_6(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T1, T2, T4, T5, T7, T8;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
T4 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
T5 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
T7 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
T8 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
{
V T3, Td, T6, Te, T9, Tf;
T3 = VSUB(T1, T2);
Td = VADD(T1, T2);
T6 = VSUB(T4, T5);
Te = VADD(T4, T5);
T9 = VSUB(T7, T8);
Tf = VADD(T7, T8);
{
V Tg, Ti, Ta, Tc;
Tg = VADD(Te, Tf);
Ti = VMUL(LDK(KP866025403), VSUB(Te, Tf));
Ta = VADD(T6, T9);
Tc = VMUL(LDK(KP866025403), VSUB(T6, T9));
{
V Th, Tj, Tb, Tk;
Th = VFNMS(LDK(KP500000000), Tg, Td);
Tj = VADD(Td, Tg);
STM2(&(xo[0]), Tj, ovs, &(xo[0]));
Tb = VFNMS(LDK(KP500000000), Ta, T3);
Tk = VADD(T3, Ta);
STM2(&(xo[6]), Tk, ovs, &(xo[2]));
{
V Tl, Tm, Tn, To;
Tl = VFMAI(Ti, Th);
STM2(&(xo[8]), Tl, ovs, &(xo[0]));
Tm = VFNMSI(Ti, Th);
STM2(&(xo[4]), Tm, ovs, &(xo[0]));
STN2(&(xo[4]), Tm, Tk, ovs);
Tn = VFNMSI(Tc, Tb);
STM2(&(xo[10]), Tn, ovs, &(xo[2]));
STN2(&(xo[8]), Tl, Tn, ovs);
To = VFMAI(Tc, Tb);
STM2(&(xo[2]), To, ovs, &(xo[2]));
STN2(&(xo[0]), Tj, To, ovs);
}
}
}
}
}
}
示例4: n2bv_10
static void n2bv_10(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP250000000, +0.250000000000000000000000000000000000000000000);
DVK(KP559016994, +0.559016994374947424102293417182819058860154590);
DVK(KP587785252, +0.587785252292473129168705954639072768597652438);
DVK(KP951056516, +0.951056516295153572116439333379382143405698634);
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V Tl, Ty, T7, Te, Tw, Tt, Tz, TA, TB, Tg, Th, Tm, Tj, Tk;
Tj = LD(&(xi[0]), ivs, &(xi[0]));
Tk = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
Tl = VSUB(Tj, Tk);
Ty = VADD(Tj, Tk);
{
V T3, Tr, Td, Tv, T6, Ts, Ta, Tu;
{
V T1, T2, Tb, Tc;
T1 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
T3 = VSUB(T1, T2);
Tr = VADD(T1, T2);
Tb = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
Tc = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
Td = VSUB(Tb, Tc);
Tv = VADD(Tb, Tc);
}
{
V T4, T5, T8, T9;
T4 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
T5 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
T6 = VSUB(T4, T5);
Ts = VADD(T4, T5);
T8 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
T9 = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
Ta = VSUB(T8, T9);
Tu = VADD(T8, T9);
}
T7 = VSUB(T3, T6);
Te = VSUB(Ta, Td);
Tw = VSUB(Tu, Tv);
Tt = VSUB(Tr, Ts);
Tz = VADD(Tr, Ts);
TA = VADD(Tu, Tv);
TB = VADD(Tz, TA);
Tg = VADD(T3, T6);
Th = VADD(Ta, Td);
Tm = VADD(Tg, Th);
}
{
V TH, TI, TK, TL, TM;
TH = VADD(Tl, Tm);
STM2(&(xo[10]), TH, ovs, &(xo[2]));
TI = VADD(Ty, TB);
STM2(&(xo[0]), TI, ovs, &(xo[0]));
{
V Tf, Tq, To, Tp, Ti, Tn, TJ;
Tf = VBYI(VFMA(LDK(KP951056516), T7, VMUL(LDK(KP587785252), Te)));
Tq = VBYI(VFNMS(LDK(KP951056516), Te, VMUL(LDK(KP587785252), T7)));
Ti = VMUL(LDK(KP559016994), VSUB(Tg, Th));
Tn = VFNMS(LDK(KP250000000), Tm, Tl);
To = VADD(Ti, Tn);
Tp = VSUB(Tn, Ti);
TJ = VADD(Tf, To);
STM2(&(xo[2]), TJ, ovs, &(xo[2]));
STN2(&(xo[0]), TI, TJ, ovs);
TK = VADD(Tq, Tp);
STM2(&(xo[14]), TK, ovs, &(xo[2]));
TL = VSUB(To, Tf);
STM2(&(xo[18]), TL, ovs, &(xo[2]));
TM = VSUB(Tp, Tq);
STM2(&(xo[6]), TM, ovs, &(xo[2]));
}
{
V Tx, TG, TE, TF, TC, TD;
Tx = VBYI(VFNMS(LDK(KP951056516), Tw, VMUL(LDK(KP587785252), Tt)));
TG = VBYI(VFMA(LDK(KP951056516), Tt, VMUL(LDK(KP587785252), Tw)));
TC = VFNMS(LDK(KP250000000), TB, Ty);
TD = VMUL(LDK(KP559016994), VSUB(Tz, TA));
TE = VSUB(TC, TD);
TF = VADD(TD, TC);
{
V TN, TO, TP, TQ;
TN = VADD(Tx, TE);
STM2(&(xo[4]), TN, ovs, &(xo[0]));
STN2(&(xo[4]), TN, TM, ovs);
TO = VADD(TG, TF);
STM2(&(xo[12]), TO, ovs, &(xo[0]));
STN2(&(xo[12]), TO, TK, ovs);
TP = VSUB(TE, Tx);
STM2(&(xo[16]), TP, ovs, &(xo[0]));
STN2(&(xo[16]), TP, TL, ovs);
TQ = VSUB(TF, TG);
STM2(&(xo[8]), TQ, ovs, &(xo[0]));
STN2(&(xo[8]), TQ, TH, ovs);
}
}
//.........这里部分代码省略.........
示例5: n2bv_12
static void n2bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
{
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T5, Ta, TG, TF, Ty, Tm, Ti, Tp, TJ, TI, Tx, Ts;
{
V T1, T6, T4, Tk, T9, Tl;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T6 = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
{
V T2, T3, T7, T8;
T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
T3 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
T4 = VADD(T2, T3);
Tk = VSUB(T2, T3);
T7 = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
T8 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
T9 = VADD(T7, T8);
Tl = VSUB(T7, T8);
}
T5 = VFNMS(LDK(KP500000000), T4, T1);
Ta = VFNMS(LDK(KP500000000), T9, T6);
TG = VADD(T6, T9);
TF = VADD(T1, T4);
Ty = VADD(Tk, Tl);
Tm = VMUL(LDK(KP866025403), VSUB(Tk, Tl));
}
{
V Tn, Tq, Te, To, Th, Tr;
Tn = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
Tq = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
{
V Tc, Td, Tf, Tg;
Tc = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
Td = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
Te = VSUB(Tc, Td);
To = VADD(Tc, Td);
Tf = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
Tg = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
Th = VSUB(Tf, Tg);
Tr = VADD(Tf, Tg);
}
Ti = VMUL(LDK(KP866025403), VSUB(Te, Th));
Tp = VFNMS(LDK(KP500000000), To, Tn);
TJ = VADD(Tq, Tr);
TI = VADD(Tn, To);
Tx = VADD(Te, Th);
Ts = VFNMS(LDK(KP500000000), Tr, Tq);
}
{
V TN, TO, TP, TQ, TR, TS;
{
V TH, TK, TL, TM;
TH = VSUB(TF, TG);
TK = VBYI(VSUB(TI, TJ));
TN = VSUB(TH, TK);
STM2(&(xo[6]), TN, ovs, &(xo[2]));
TO = VADD(TH, TK);
STM2(&(xo[18]), TO, ovs, &(xo[2]));
TL = VADD(TF, TG);
TM = VADD(TI, TJ);
TP = VSUB(TL, TM);
STM2(&(xo[12]), TP, ovs, &(xo[0]));
TQ = VADD(TL, TM);
STM2(&(xo[0]), TQ, ovs, &(xo[0]));
}
{
V Tj, Tv, Tu, Tw, Tb, Tt, TT, TU;
Tb = VSUB(T5, Ta);
Tj = VSUB(Tb, Ti);
Tv = VADD(Tb, Ti);
Tt = VSUB(Tp, Ts);
Tu = VBYI(VADD(Tm, Tt));
Tw = VBYI(VSUB(Tt, Tm));
TR = VSUB(Tj, Tu);
STM2(&(xo[22]), TR, ovs, &(xo[2]));
TS = VADD(Tv, Tw);
STM2(&(xo[10]), TS, ovs, &(xo[2]));
TT = VADD(Tj, Tu);
STM2(&(xo[2]), TT, ovs, &(xo[2]));
STN2(&(xo[0]), TQ, TT, ovs);
TU = VSUB(Tv, Tw);
STM2(&(xo[14]), TU, ovs, &(xo[2]));
STN2(&(xo[12]), TP, TU, ovs);
}
{
V Tz, TD, TC, TE, TA, TB;
Tz = VBYI(VMUL(LDK(KP866025403), VSUB(Tx, Ty)));
TD = VBYI(VMUL(LDK(KP866025403), VADD(Ty, Tx)));
TA = VADD(T5, Ta);
TB = VADD(Tp, Ts);
TC = VSUB(TA, TB);
TE = VADD(TA, TB);
//.........这里部分代码省略.........
示例6: n2bv_14
static void n2bv_14(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP900968867, +0.900968867902419126236102319507445051165919162);
DVK(KP801937735, +0.801937735804838252472204639014890102331838324);
DVK(KP974927912, +0.974927912181823607018131682993931217232785801);
DVK(KP692021471, +0.692021471630095869627814897002069140197260599);
DVK(KP554958132, +0.554958132087371191422194871006410481067288862);
DVK(KP356895867, +0.356895867892209443894399510021300583399127187);
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V TH, T3, TP, Tn, Ta, Tu, TU, TK, TO, Tk, TM, Tg, TL, Td, T1;
V T2;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
{
V Ti, TI, T6, TJ, T9, Tj, Te, Tf, Tb, Tc;
{
V T4, T5, T7, T8, Tl, Tm;
T4 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
T5 = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
T7 = LD(&(xi[WS(is, 12)]), ivs, &(xi[0]));
T8 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
Tl = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
Tm = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
Ti = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
TH = VADD(T1, T2);
T3 = VSUB(T1, T2);
TI = VADD(T4, T5);
T6 = VSUB(T4, T5);
TJ = VADD(T7, T8);
T9 = VSUB(T7, T8);
TP = VADD(Tl, Tm);
Tn = VSUB(Tl, Tm);
Tj = LD(&(xi[WS(is, 13)]), ivs, &(xi[WS(is, 1)]));
Te = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
Tf = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
Tb = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
Tc = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
}
Ta = VADD(T6, T9);
Tu = VSUB(T6, T9);
TU = VSUB(TI, TJ);
TK = VADD(TI, TJ);
TO = VADD(Ti, Tj);
Tk = VSUB(Ti, Tj);
TM = VADD(Te, Tf);
Tg = VSUB(Te, Tf);
TL = VADD(Tb, Tc);
Td = VSUB(Tb, Tc);
}
{
V T19, T1a, T13, TG, TY, T18, TB, Tw, TT, Tz, T11, T16, TE, Tr, TV;
V TQ;
TV = VSUB(TP, TO);
TQ = VADD(TO, TP);
{
V Ts, To, TW, TN;
Ts = VSUB(Tk, Tn);
To = VADD(Tk, Tn);
TW = VSUB(TM, TL);
TN = VADD(TL, TM);
{
V Tt, Th, TR, T12;
Tt = VSUB(Td, Tg);
Th = VADD(Td, Tg);
TR = VFNMS(LDK(KP356895867), TK, TQ);
T12 = VFNMS(LDK(KP554958132), TV, TU);
{
V Tx, TF, TZ, T14;
Tx = VFNMS(LDK(KP356895867), Ta, To);
TF = VFMA(LDK(KP554958132), Ts, Tu);
T19 = VADD(TH, VADD(TK, VADD(TN, TQ)));
STM2(&(xo[0]), T19, ovs, &(xo[0]));
TZ = VFNMS(LDK(KP356895867), TN, TK);
T14 = VFNMS(LDK(KP356895867), TQ, TN);
{
V TX, T17, TC, Tp;
TX = VFMA(LDK(KP554958132), TW, TV);
T17 = VFMA(LDK(KP554958132), TU, TW);
T1a = VADD(T3, VADD(Ta, VADD(Th, To)));
STM2(&(xo[14]), T1a, ovs, &(xo[2]));
TC = VFNMS(LDK(KP356895867), Th, Ta);
Tp = VFNMS(LDK(KP356895867), To, Th);
{
V TA, Tv, TS, Ty;
TA = VFMA(LDK(KP554958132), Tt, Ts);
Tv = VFNMS(LDK(KP554958132), Tu, Tt);
TS = VFNMS(LDK(KP692021471), TR, TN);
T13 = VMUL(LDK(KP974927912), VFNMS(LDK(KP801937735), T12, TW));
Ty = VFNMS(LDK(KP692021471), Tx, Th);
TG = VMUL(LDK(KP974927912), VFMA(LDK(KP801937735), TF, Tt));
{
V T10, T15, TD, Tq;
T10 = VFNMS(LDK(KP692021471), TZ, TQ);
T15 = VFNMS(LDK(KP692021471), T14, TK);
TY = VMUL(LDK(KP974927912), VFMA(LDK(KP801937735), TX, TU));
//.........这里部分代码省略.........
示例7: n2fv_16
//.........这里部分代码省略.........
Tk = VADD(Ti, Tj);
TW = VFNMS(LDK(KP707106781), TV, TU);
T16 = VFMA(LDK(KP707106781), TV, TU);
}
{
V TY, TR, Tl, TD;
TY = VFMA(LDK(KP414213562), TP, TQ);
TR = VFNMS(LDK(KP414213562), TQ, TP);
Tl = VSUB(Th, Tk);
TD = VADD(Th, Tk);
{
V TS, T17, TZ, T14;
TS = VADD(TO, TR);
T17 = VSUB(TR, TO);
TZ = VSUB(TX, TY);
T14 = VADD(TX, TY);
{
V TE, TG, Tm, Tv;
TE = VADD(TC, TD);
TG = VSUB(TD, TC);
Tm = VADD(Te, Tl);
Tv = VSUB(Tl, Te);
{
V T18, T1a, TT, T11;
T18 = VFNMS(LDK(KP923879532), T17, T16);
T1a = VFMA(LDK(KP923879532), T17, T16);
TT = VFNMS(LDK(KP923879532), TS, TL);
T11 = VFMA(LDK(KP923879532), TS, TL);
{
V T15, T19, T10, T12;
T15 = VFNMS(LDK(KP923879532), T14, T13);
T19 = VFMA(LDK(KP923879532), T14, T13);
T10 = VFNMS(LDK(KP923879532), TZ, TW);
T12 = VFMA(LDK(KP923879532), TZ, TW);
{
V T1b, T1c, T1d, T1e;
T1b = VFMAI(TG, TF);
STM2(&(xo[8]), T1b, ovs, &(xo[0]));
T1c = VFNMSI(TG, TF);
STM2(&(xo[24]), T1c, ovs, &(xo[0]));
T1d = VADD(TB, TE);
STM2(&(xo[0]), T1d, ovs, &(xo[0]));
T1e = VSUB(TB, TE);
STM2(&(xo[16]), T1e, ovs, &(xo[0]));
{
V Tw, Ty, Tn, Tx;
Tw = VFNMS(LDK(KP707106781), Tv, Tu);
Ty = VFMA(LDK(KP707106781), Tv, Tu);
Tn = VFNMS(LDK(KP707106781), Tm, T7);
Tx = VFMA(LDK(KP707106781), Tm, T7);
{
V T1f, T1g, T1h, T1i;
T1f = VFMAI(T1a, T19);
STM2(&(xo[6]), T1f, ovs, &(xo[2]));
T1g = VFNMSI(T1a, T19);
STM2(&(xo[26]), T1g, ovs, &(xo[2]));
STN2(&(xo[24]), T1c, T1g, ovs);
T1h = VFMAI(T18, T15);
STM2(&(xo[22]), T1h, ovs, &(xo[2]));
T1i = VFNMSI(T18, T15);
STM2(&(xo[10]), T1i, ovs, &(xo[2]));
STN2(&(xo[8]), T1b, T1i, ovs);
{
V T1j, T1k, T1l, T1m;
T1j = VFNMSI(T12, T11);
STM2(&(xo[2]), T1j, ovs, &(xo[2]));
STN2(&(xo[0]), T1d, T1j, ovs);
T1k = VFMAI(T12, T11);
STM2(&(xo[30]), T1k, ovs, &(xo[2]));
T1l = VFMAI(T10, TT);
STM2(&(xo[14]), T1l, ovs, &(xo[2]));
T1m = VFNMSI(T10, TT);
STM2(&(xo[18]), T1m, ovs, &(xo[2]));
STN2(&(xo[16]), T1e, T1m, ovs);
{
V T1n, T1o, T1p, T1q;
T1n = VFNMSI(Ty, Tx);
STM2(&(xo[28]), T1n, ovs, &(xo[0]));
STN2(&(xo[28]), T1n, T1k, ovs);
T1o = VFMAI(Ty, Tx);
STM2(&(xo[4]), T1o, ovs, &(xo[0]));
STN2(&(xo[4]), T1o, T1f, ovs);
T1p = VFMAI(Tw, Tn);
STM2(&(xo[20]), T1p, ovs, &(xo[0]));
STN2(&(xo[20]), T1p, T1h, ovs);
T1q = VFNMSI(Tw, Tn);
STM2(&(xo[12]), T1q, ovs, &(xo[0]));
STN2(&(xo[12]), T1q, T1l, ovs);
}
}
}
}
}
}
}
}
}
}
}
}
示例8: n2bv_32
//.........这里部分代码省略.........
T2J = VADD(T29, T2a);
Ti = VFMA(LDK(KP382683432), Te, VMUL(LDK(KP923879532), Th));
Tp = VFNMS(LDK(KP382683432), To, VMUL(LDK(KP923879532), Tl));
Tq = VSUB(Ti, Tp);
T1A = VADD(Ti, Tp);
{
V T12, T13, T28, T2b;
T12 = VFNMS(LDK(KP382683432), Th, VMUL(LDK(KP923879532), Te));
T13 = VFMA(LDK(KP923879532), To, VMUL(LDK(KP382683432), Tl));
T14 = VSUB(T12, T13);
T1q = VADD(T12, T13);
T28 = VSUB(T26, T27);
T2b = VSUB(T29, T2a);
T2c = VMUL(LDK(KP707106781), VSUB(T28, T2b));
T2l = VMUL(LDK(KP707106781), VADD(T28, T2b));
}
}
{
V T31, T32, T33, T34, T35, T36, T37, T38, T39, T3a, T3b, T3c;
{
V T2L, T2R, T2Q, T2S;
{
V T2H, T2K, T2O, T2P;
T2H = VMUL(LDK(KP707106781), VSUB(T2D, T2G));
T2K = VSUB(T2I, T2J);
T2L = VBYI(VSUB(T2H, T2K));
T2R = VBYI(VADD(T2K, T2H));
T2O = VSUB(T2M, T2N);
T2P = VMUL(LDK(KP707106781), VADD(T2D, T2G));
T2Q = VSUB(T2O, T2P);
T2S = VADD(T2O, T2P);
}
T31 = VADD(T2L, T2Q);
STM2(&(xo[24]), T31, ovs, &(xo[0]));
T32 = VSUB(T2S, T2R);
STM2(&(xo[56]), T32, ovs, &(xo[0]));
T33 = VSUB(T2Q, T2L);
STM2(&(xo[40]), T33, ovs, &(xo[0]));
T34 = VADD(T2R, T2S);
STM2(&(xo[8]), T34, ovs, &(xo[0]));
}
{
V T2h, T2r, T2q, T2s;
{
V T25, T2g, T2m, T2p;
T25 = VSUB(T1X, T24);
T2g = VSUB(T2c, T2f);
T2h = VBYI(VSUB(T25, T2g));
T2r = VBYI(VADD(T2g, T25));
T2m = VSUB(T2k, T2l);
T2p = VSUB(T2n, T2o);
T2q = VSUB(T2m, T2p);
T2s = VADD(T2m, T2p);
}
T35 = VADD(T2h, T2q);
STM2(&(xo[20]), T35, ovs, &(xo[0]));
T36 = VSUB(T2s, T2r);
STM2(&(xo[52]), T36, ovs, &(xo[0]));
T37 = VSUB(T2q, T2h);
STM2(&(xo[44]), T37, ovs, &(xo[0]));
T38 = VADD(T2r, T2s);
STM2(&(xo[12]), T38, ovs, &(xo[0]));
}
{
V T2V, T2Z, T2Y, T30;
{
示例9: n2fv_20
//.........这里部分代码省略.........
TG = VSUB(Ts, Tp);
TN = VSUB(T6, T9);
TO = VSUB(Td, Tg);
TH = VSUB(Tz, Tw);
T13 = VSUB(T11, T12);
T16 = VSUB(T14, T15);
T1k = VADD(T13, T16);
T1u = VADD(T11, T12);
T1v = VADD(T14, T15);
T1z = VADD(T1u, T1v);
T1r = VADD(T18, T19);
T1s = VADD(T1b, T1c);
T1y = VADD(T1r, T1s);
T1a = VSUB(T18, T19);
T1d = VSUB(T1b, T1c);
T1j = VADD(T1a, T1d);
{
V Ta, Th, Tt, TA;
Ta = VADD(T6, T9);
Th = VADD(Td, Tg);
Ti = VADD(Ta, Th);
TD = VMUL(LDK(KP559016994), VSUB(Ta, Th));
Tt = VADD(Tp, Ts);
TA = VADD(Tw, Tz);
TB = VADD(Tt, TA);
TL = VMUL(LDK(KP559016994), VSUB(TA, Tt));
}
}
{
V T1I, T1J, T1K, T1L, T1N, T1H, Tj, TC;
Tj = VADD(T3, Ti);
TC = VBYI(VADD(Tm, TB));
T1H = VSUB(Tj, TC);
STM2(&(xo[10]), T1H, ovs, &(xo[2]));
T1I = VADD(Tj, TC);
STM2(&(xo[30]), T1I, ovs, &(xo[2]));
{
V T1A, T1C, T1D, T1x, T1G, T1t, T1w, T1F, T1E, T1M;
T1A = VMUL(LDK(KP559016994), VSUB(T1y, T1z));
T1C = VADD(T1y, T1z);
T1D = VFNMS(LDK(KP250000000), T1C, T1B);
T1t = VSUB(T1r, T1s);
T1w = VSUB(T1u, T1v);
T1x = VBYI(VFMA(LDK(KP951056516), T1t, VMUL(LDK(KP587785252), T1w)));
T1G = VBYI(VFNMS(LDK(KP587785252), T1t, VMUL(LDK(KP951056516), T1w)));
T1J = VADD(T1B, T1C);
STM2(&(xo[0]), T1J, ovs, &(xo[0]));
T1F = VSUB(T1D, T1A);
T1K = VSUB(T1F, T1G);
STM2(&(xo[16]), T1K, ovs, &(xo[0]));
T1L = VADD(T1G, T1F);
STM2(&(xo[24]), T1L, ovs, &(xo[0]));
T1E = VADD(T1A, T1D);
T1M = VADD(T1x, T1E);
STM2(&(xo[8]), T1M, ovs, &(xo[0]));
STN2(&(xo[8]), T1M, T1H, ovs);
T1N = VSUB(T1E, T1x);
STM2(&(xo[32]), T1N, ovs, &(xo[0]));
}
{
V T1O, T1P, T1R, T1S;
{
V T1n, T1l, T1m, T1f, T1q, T17, T1e, T1p, T1Q, T1o;
T1n = VMUL(LDK(KP559016994), VSUB(T1j, T1k));
T1l = VADD(T1j, T1k);
T1m = VFNMS(LDK(KP250000000), T1l, T1i);
示例10: n2bv_8
static void n2bv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T1, T2, Tc, Td, T4, T5, T7, T8;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
Tc = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
Td = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
T4 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
T5 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
T7 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
T8 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
{
V T3, Tj, Te, Tk, T6, Tm, T9, Tn, Tp, Tl;
T3 = VSUB(T1, T2);
Tj = VADD(T1, T2);
Te = VSUB(Tc, Td);
Tk = VADD(Tc, Td);
T6 = VSUB(T4, T5);
Tm = VADD(T4, T5);
T9 = VSUB(T7, T8);
Tn = VADD(T7, T8);
Tp = VADD(Tj, Tk);
Tl = VSUB(Tj, Tk);
{
V Tq, To, Ta, Tf;
Tq = VADD(Tm, Tn);
To = VSUB(Tm, Tn);
Ta = VADD(T6, T9);
Tf = VSUB(T6, T9);
{
V Tr, Ts, Tt, Tu, Tg, Ti, Tb, Th;
Tr = VFMAI(To, Tl);
STM2(&(xo[4]), Tr, ovs, &(xo[0]));
Ts = VFNMSI(To, Tl);
STM2(&(xo[12]), Ts, ovs, &(xo[0]));
Tt = VADD(Tp, Tq);
STM2(&(xo[0]), Tt, ovs, &(xo[0]));
Tu = VSUB(Tp, Tq);
STM2(&(xo[8]), Tu, ovs, &(xo[0]));
Tg = VFNMS(LDK(KP707106781), Tf, Te);
Ti = VFMA(LDK(KP707106781), Tf, Te);
Tb = VFNMS(LDK(KP707106781), Ta, T3);
Th = VFMA(LDK(KP707106781), Ta, T3);
{
V Tv, Tw, Tx, Ty;
Tv = VFNMSI(Ti, Th);
STM2(&(xo[14]), Tv, ovs, &(xo[2]));
STN2(&(xo[12]), Ts, Tv, ovs);
Tw = VFMAI(Ti, Th);
STM2(&(xo[2]), Tw, ovs, &(xo[2]));
STN2(&(xo[0]), Tt, Tw, ovs);
Tx = VFMAI(Tg, Tb);
STM2(&(xo[10]), Tx, ovs, &(xo[2]));
STN2(&(xo[8]), Tu, Tx, ovs);
Ty = VFNMSI(Tg, Tb);
STM2(&(xo[6]), Ty, ovs, &(xo[2]));
STN2(&(xo[4]), Tr, Ty, ovs);
}
}
}
}
}
}
示例11: n2fv_8
static void n2fv_8(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP707106781, +0.707106781186547524400844362104849039284835938);
{
INT i;
const R *xi;
R *xo;
xi = ri;
xo = ro;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T3, Tj, Tf, Tk, Ta, Tn, Tc, Tm, Ts, Tu;
{
V T1, T2, Td, Te;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
T3 = VSUB(T1, T2);
Tj = VADD(T1, T2);
Td = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
Te = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
Tf = VSUB(Td, Te);
Tk = VADD(Td, Te);
{
V T4, T5, T6, T7, T8, T9;
T4 = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
T5 = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
T6 = VSUB(T4, T5);
T7 = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
T8 = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
T9 = VSUB(T7, T8);
Ta = VMUL(LDK(KP707106781), VADD(T6, T9));
Tn = VADD(T7, T8);
Tc = VMUL(LDK(KP707106781), VSUB(T9, T6));
Tm = VADD(T4, T5);
}
}
{
V Tr, Tb, Tg, Tp, Tq, Tt;
Tb = VADD(T3, Ta);
Tg = VBYI(VSUB(Tc, Tf));
Tr = VSUB(Tb, Tg);
STM2(&(xo[14]), Tr, ovs, &(xo[2]));
Ts = VADD(Tb, Tg);
STM2(&(xo[2]), Ts, ovs, &(xo[2]));
Tp = VSUB(Tj, Tk);
Tq = VBYI(VSUB(Tn, Tm));
Tt = VSUB(Tp, Tq);
STM2(&(xo[12]), Tt, ovs, &(xo[0]));
STN2(&(xo[12]), Tt, Tr, ovs);
Tu = VADD(Tp, Tq);
STM2(&(xo[4]), Tu, ovs, &(xo[0]));
}
{
V Tv, Th, Ti, Tw;
Th = VSUB(T3, Ta);
Ti = VBYI(VADD(Tf, Tc));
Tv = VSUB(Th, Ti);
STM2(&(xo[10]), Tv, ovs, &(xo[2]));
Tw = VADD(Th, Ti);
STM2(&(xo[6]), Tw, ovs, &(xo[2]));
STN2(&(xo[4]), Tu, Tw, ovs);
{
V Tl, To, Tx, Ty;
Tl = VADD(Tj, Tk);
To = VADD(Tm, Tn);
Tx = VSUB(Tl, To);
STM2(&(xo[8]), Tx, ovs, &(xo[0]));
STN2(&(xo[8]), Tx, Tv, ovs);
Ty = VADD(Tl, To);
STM2(&(xo[0]), Ty, ovs, &(xo[0]));
STN2(&(xo[0]), Ty, Ts, ovs);
}
}
}
}
VLEAVE();
}
示例12: n2bv_12
static void n2bv_12(const R *ri, const R *ii, R *ro, R *io, stride is, stride os, INT v, INT ivs, INT ovs)
{
DVK(KP866025403, +0.866025403784438646763723170752936183471402627);
DVK(KP500000000, +0.500000000000000000000000000000000000000000000);
INT i;
const R *xi;
R *xo;
xi = ii;
xo = io;
for (i = v; i > 0; i = i - VL, xi = xi + (VL * ivs), xo = xo + (VL * ovs), MAKE_VOLATILE_STRIDE(is), MAKE_VOLATILE_STRIDE(os)) {
V T1, T6, Tc, Th, Td, Te, Ti, Tz, T4, TA, T9, Tj, Tf, Tw;
{
V T2, T3, T7, T8;
T1 = LD(&(xi[0]), ivs, &(xi[0]));
T6 = LD(&(xi[WS(is, 6)]), ivs, &(xi[0]));
T2 = LD(&(xi[WS(is, 4)]), ivs, &(xi[0]));
T3 = LD(&(xi[WS(is, 8)]), ivs, &(xi[0]));
T7 = LD(&(xi[WS(is, 10)]), ivs, &(xi[0]));
T8 = LD(&(xi[WS(is, 2)]), ivs, &(xi[0]));
Tc = LD(&(xi[WS(is, 3)]), ivs, &(xi[WS(is, 1)]));
Th = LD(&(xi[WS(is, 9)]), ivs, &(xi[WS(is, 1)]));
Td = LD(&(xi[WS(is, 7)]), ivs, &(xi[WS(is, 1)]));
Te = LD(&(xi[WS(is, 11)]), ivs, &(xi[WS(is, 1)]));
Ti = LD(&(xi[WS(is, 1)]), ivs, &(xi[WS(is, 1)]));
Tz = VSUB(T2, T3);
T4 = VADD(T2, T3);
TA = VSUB(T7, T8);
T9 = VADD(T7, T8);
Tj = LD(&(xi[WS(is, 5)]), ivs, &(xi[WS(is, 1)]));
}
Tf = VADD(Td, Te);
Tw = VSUB(Td, Te);
{
V T5, Tp, TJ, TB, Ta, Tq, Tk, Tx, Tg, Ts;
T5 = VADD(T1, T4);
Tp = VFNMS(LDK(KP500000000), T4, T1);
TJ = VSUB(Tz, TA);
TB = VADD(Tz, TA);
Ta = VADD(T6, T9);
Tq = VFNMS(LDK(KP500000000), T9, T6);
Tk = VADD(Ti, Tj);
Tx = VSUB(Tj, Ti);
Tg = VADD(Tc, Tf);
Ts = VFNMS(LDK(KP500000000), Tf, Tc);
{
V Tr, TF, Tb, Tn, TG, Ty, Tl, Tt;
Tr = VADD(Tp, Tq);
TF = VSUB(Tp, Tq);
Tb = VSUB(T5, Ta);
Tn = VADD(T5, Ta);
TG = VADD(Tw, Tx);
Ty = VSUB(Tw, Tx);
Tl = VADD(Th, Tk);
Tt = VFNMS(LDK(KP500000000), Tk, Th);
{
V TC, TE, TH, TL, Tu, TI, Tm, To;
TC = VMUL(LDK(KP866025403), VSUB(Ty, TB));
TE = VMUL(LDK(KP866025403), VADD(TB, Ty));
TH = VFNMS(LDK(KP866025403), TG, TF);
TL = VFMA(LDK(KP866025403), TG, TF);
Tu = VADD(Ts, Tt);
TI = VSUB(Ts, Tt);
Tm = VSUB(Tg, Tl);
To = VADD(Tg, Tl);
{
V TK, TM, Tv, TD;
TK = VFMA(LDK(KP866025403), TJ, TI);
TM = VFNMS(LDK(KP866025403), TJ, TI);
Tv = VSUB(Tr, Tu);
TD = VADD(Tr, Tu);
{
V TN, TO, TP, TQ;
TN = VADD(Tn, To);
STM2(&(xo[0]), TN, ovs, &(xo[0]));
TO = VSUB(Tn, To);
STM2(&(xo[12]), TO, ovs, &(xo[0]));
TP = VFMAI(Tm, Tb);
STM2(&(xo[18]), TP, ovs, &(xo[2]));
TQ = VFNMSI(Tm, Tb);
STM2(&(xo[6]), TQ, ovs, &(xo[2]));
{
V TR, TS, TT, TU;
TR = VFMAI(TM, TL);
STM2(&(xo[10]), TR, ovs, &(xo[2]));
TS = VFNMSI(TM, TL);
STM2(&(xo[14]), TS, ovs, &(xo[2]));
STN2(&(xo[12]), TO, TS, ovs);
TT = VFNMSI(TK, TH);
STM2(&(xo[22]), TT, ovs, &(xo[2]));
TU = VFMAI(TK, TH);
STM2(&(xo[2]), TU, ovs, &(xo[2]));
STN2(&(xo[0]), TN, TU, ovs);
{
V TV, TW, TX, TY;
TV = VFNMSI(TE, TD);
STM2(&(xo[16]), TV, ovs, &(xo[0]));
STN2(&(xo[16]), TV, TP, ovs);
TW = VFMAI(TE, TD);
STM2(&(xo[8]), TW, ovs, &(xo[0]));
STN2(&(xo[8]), TW, TR, ovs);
//.........这里部分代码省略.........